O’Horo, John C.

Abstract Background The use of extracorporeal membrane oxygenation (ECMO) in critically ill adults is increasing. Patients on ECMO are at high risk for infections, with 20.5% of adults acquiring infections while on ECMO. An Extracorporeal Life Support Organization (ELSO) Infectious Disease Task Force statement concluded that no antibiotic prophylaxis is needed for patients on ECMO though it also noted that this was based on limited data. We implemented an antimicrobial prophylaxis protocol for patients on ECMO at our institution and analyzed antimicrobial use and outcomes in these patients with a pre- and post-analysis. Methods We conducted a retrospective review of 294 patients on ECMO between July 1, 2011 and July 1, 2017. An ECMO antimicrobial prophylaxis guideline was initially implemented on July 1, 2014; there was poor adherence to the guideline and antimicrobial use actually increased. A more restrictive protocol was implemented in November 2018 with input from stakeholders including cardiac surgeons, critical care and infectious disease (ID) providers. We had a cohort of 161 patients before (July 2014–November 2018) and 37 patients after (November 2018–April 2018) the implementation of the updated protocol. We evaluated primary outcomes of gross days of antimicrobial use, percent of antibiotic-free days and days of individual antimicrobial use, adjusted for APACHE scores and ECMO duration. Results When adjusted for days on ECMO, mean antibiotic days decreased after implementation of the protocol; for vancomycin (0.27 vs. 0.02, P < 0.0003), cefepime (0.15 vs. 0.02, P < 0.02), meropenem (0.09 vs. 0, P < 0.02), zosyn (0.16 vs. 0, P < 0.002), caspofungin (0.346, 0.138 P < 0.003). This was accompanied by a nonsignificant increase in mean fluconazole use (0.29 vs. 0.37, P < 0.3). There was no impact on patient mortality or nosocomial infection rate. Additional results can be found in table. Conclusion The use of an antimicrobial prophylaxis protocol in ECMO patients led to improvement in antimicrobial usage without increasing nosocomial infections in a population at a high risk of infection. Disclosures All authors: No reported disclosures.

Abstract Background Infectious diseases (ID) consultation is associated with improved outcomes for patients hospitalized with infection. Provision of this service is not practical for many smaller hospitals, particularly critical access facilities. We sought to evaluate the impact of an electronic, asynchronous consultative service (“eConsults”) at two rural hospitals without on-site ID support. Methods Beginning July 2018, ID physicians at Mayo Clinic, Rochester, began performing asynchronous eConsults for patients hospitalized at Mayo Clinic Health System (MCHS) Albert Lea and MCHS Austin Hospitals. The first 100 consecutive patients receiving eConsults (cases) were compared with patients admitted for infection at the same facilities prior to pilot initiation (controls). Cases were matched to controls using 1:3 propensity match based on age, gender, race, and weighted Charlson comorbidity index. The primary outcome was readmission or death within 30 days after hospital discharge. Results Cases (n = 100) were more likely to have been hospitalized in the 6 months prior to the index hospitalization than controls (n = 300) (P < 0.0001). Patients with ID eConsult had a significantly decreased odds of death at 30 days (OR 0.3, 95% CI 0.2–0.7, P = 0.003) and there was a trend toward decreased readmission at 30 days (OR 0.4, 95% CI 0.2–1.1, P = 0.07). The mean length of stay was significantly longer for cases (5.7 days vs. 3.8 days, P =0.003). However, ID eConsult did not occur until 2.6 days into hospitalization, on average. Conclusion ID eConsultation was associated with lower odds of 30-day mortality. The increased length of stay may be mitigated by encouraging ID eConsults earlier during hospitalization. Inpatient ID eConsults represent a high-value proposition for patients and providers and a feasible means of expanding the reach of ID physicians. This novel care delivery model warrants further investigations. Disclosures All authors: No reported disclosures.

Abstract Background Demographic and epidemiologic characteristics of Hepatitis E virus (HEV) infected patients in the United States are not well-described. HEV infection may result in severe complications and lead to chronic infection and cirrhosis, especially in immunocompromised patients. There are no widely accepted guidelines for HEV screening and testing in the United States. Identifying traits of known seropositive patients and comorbidities may inform better screening and prevention strategies. In this study, we describe rates of liver disease, transplant status, chronic kidney diseases, and diabetes mellitus among patients serologically tested for HEV at our institution. Methods We retrospectively reviewed all patients for whom HEV IgM or IgG serologic testing was performed across the Mayo Clinic enterprise using the Advanced Cohort Explorer tool. For patients with any documented HEV serologic test, we abstracted baseline patient characteristics and underlying comorbidities at the time of testing. We then grouped subjects according to serologic testing results by antibody type. Survival at one year from date of testing was also assessed. Results A total of 979 unique subjects were identified. The preponderance of subjects was Caucasian (781, 79.8%). Of subjects tested for HEV with serology, 123 (12.6%) had any positive serology. Breakdown of baseline characteristics and selected comorbidities are summarized in Table 1. The largest proportion of subjects, 458 (46.8%), were both IgG and IgM negative and 432 subjects received only IgM or IgG testing. Liver disease was more prevalent in patients with positive vs. negative testing (5.8% vs. 2.1%) as was higher age (average 55.1 years vs. 52.2). One-year survival was similar in all groups. Conclusion HEV serology is not commonly tested. Among those tested, seropositivity is uncommon. Our data show higher HEV seropositivity in older adults, which may represent increase risk of exposure over time. Higher percentage of positive testing was also observed in subjects with liver disease, which may indicate a possible etiologic association. Further population-based studies are needed to estimate prevalence of HEV infection and associated liver disease, outcomes in infected patients, and indications for testing in at-risk populations. Disclosures All authors: No reported disclosures.

Abstract Background Bronchoalveolar lavage (BAL) is a widely used procedure in the diagnosis of pneumonia in critically ill and immunocompromised hosts. Fungal smears and cultures are commonly performed on these samples. We evaluated the yield of various fungi, including but not limited to Candida species, Aspergillus species, Penicillium species, isolated from BAL specimens at our institution to determine the yield of this test and its impact on decision making. Methods We identified adult immunocompromised patients who underwent “Bronchoscopy with Immunocompromised Host Protocol (ICH),” which consists of an exhaustive list of diagnostic tests for various pathogenic organisms, over a one year period from January 1, 2017 to December 31, 2017. We reviewed if positive fungal cultures led to a change in management and if this was appropriate. Results 582 patients underwent bronchoscopy with ICH protocol. There were 285/582 (48.9%) positive fungal cultures of which 177 (62%) grew Candida species. The most common species was Candida albicans (142/177, 80%). 53(18%) were Aspergillus species of which Aspergillus fumigatus was the most common (26/53). 16/285 (5.6%) patients underwent intervention based on the results, 14(87.5%) of which were appropriate. 176/177 (99.4%) patients with Candida species in BAL cultures were not treated.10/53 (18. 8%) patients with Aspergillus species in BAL cultures were treated of which 80% were appropriate interventions based on proven/probable invasive fungal infections criteria as were rest of the 6/16 patients with other fungal organisms (Table 4). Patients with Aspergillus species in BAL cultures are 8 times more likely to have an intervention (OR: 8. 7, P = < 0. 0001) while patients with Candida species in BAL cultures are not likely to be intervened upon (OR: 0. 26, P = 0. 0098) (Table 3). Conclusion Although Candida species is commonly isolated in BAL cultures its clinical significance is minimal in the absence of disseminated disease even in immunosuppressed hosts. Evaluating the way that Candida cultures are communicated for respiratory specimens, along with diagnostic stewardship may be a route for antimicrobial stewardship. Consulting ID service early on is essential in assessing the significance of fungal culture data thereby resulting in appropriate changes in management. Disclosures All authors: No reported disclosures.

Abstract Background The interaction between influenza activity and environemental factor such as ultraviolet light index (UVI), atmospheric ozone (AO), and other related meteorological variables remains poorly understood. In the midst of climate change and increasingly poor performance of influenza vaccination, more information on influenza activity and its interaction with meteorological variables is critically needed. Methods Influenza A and B tests results by PCR from respiratory sources were collected from two large hospitals in Phoenix, AZ and Jacksonville, FL from January 1, 2014 to December 31, 2017. Publicly available meteorological data for each location was obtained from the National Oceanic and Atmospheric Administration. We excluded cases residing beyond 0.5° of longitude and latitude radius of the given meteorological data. A weekly index activity and maximum weekly values of meteorological variables were matched, and performed a correlation and regression analysis. Results A total of 5,238 influenza tests were performed during the study period. The influenza index showed an statistically significant weakly positive correlation with maximum CSUVI (r = 0.14; P = 0.0227) and mean zenith (r = 0.17; P = 0.0047). An statistically significant, positive correlation was observed between influenza index and atmospheric ozone (r = 0.23; P = 0.0001). Significant negative correlations were also observed with DBT, DPT, RH and HI (r = −0.27, r = −0.39, r = −0.13, r = −0.33, respectively; P < 0.04). The influenza index showed significant interactions in a univariate linear regression (Table 1). A relationship between influenza index and dew point temperature was observed in a multivariate model (OR = 0.66; CI95% 0.44–0.97). Conclusion To the best of our knowledge, this is the first report showing a significant interactions between influenza index, UVI and atmospheric ozone in two geographically distant locations. Further studies are needed to define the role of complex climatological patterns and influenza. Disclosures All authors: No reported disclosures.

Abstract Background Brevibacterium species are non-motile, catalase-positive, obligate aerobic gram-positive bacilli. Colonies are yellow to gray-white, non-hemolytic, smooth, 6.5% sodium chloride tolerant. B. fermentans post neurosurgical meningitis was first described in 1969 in an infant. B. casei remains the most commonly isolated species (Table 4). The most commonly reported syndromes are bloodstream infections (BSIs) and endocarditis. Despite these reports, this organism continues to be listed on CDC’s NHSN commensal database. Methods Isolates of Brevibacterium from clinical samples at Mayo Clinic, Rochester from January 1, 2014 to December 31, 2018 were identified. Charts were reviewed to determine patient demographics, immune status, source of culture, comorbidities, antibiotic susceptibility test (AST), length of stay (LOS) in hospital and intensive care unit (ICU), and mortality. Likelihood ratio (L-R) and Pearson correlation coefficient (PCC) of nominal data were calculated using the Chi-square test and Fischer exact test (FET). We defined statistical significance as P ≤ 0.05. Results We identified 48 isolates from 45 unique patients, 46% were females. Distribution of age, hospital and ICU LOS, and time to culture growth, and AST data are shown in Table 1. 15.5% patients received allogeneic or autologous stem cell (SCT), or solid-organ transplant (SOT) recipients. 89% cultures were from sterile sources and 68.75% were blood cultures. Of these, 63.64% were monomicrobial. 62% of isolates identified to species level were B. casei. 5 patients were treated; an additional 10 received active antibiotics for other indications. Statistically significant variables are reported in Tables 2 and 3. Thirty-day mortality was 13%. This was higher in patients with bacteremia (L-R: 5.3 [P = 0.02]) but FET was not statistically significance (P = 0.15). Conclusion Accurate diagnosis of Brevibacterium may require molecular techniques. At our center, SCT or SOT recipient status and recent chemotherapy were associated with bacteremia. In these patients, this organism could represent an opportunistic cause of BSI. AST data suggest that Vancomycin offers a reasonable empiric treatment option. Additional data are needed to further define host populations in whom this organism presents pathogenicity. Disclosures All authors: No reported disclosures.

Abstract Background Sepsis frequently leads to acute kidney injury. In severe cases, patients may require continuous renal replacement therapy (CRRT) which involves placement of a dialysis catheter and an extracorporeal blood filtration circuit. CRRT is commonly considered to “mask” fever, though this phenomenon has not been investigated. Methods We queried an institutional database of all patients on CRRT from 2007 to 2015 for inpatient temperature data and antibiotic administration records. Receipts of piperacillin–tazobactam, a carbapenem, or a third or fourth-generation cephalosporin, indicating a serious infection, were considered intervention arm. We analyzed temperatures recorded in the intensive care unit before, during, and after CRRT. Patients were divided into groups that did not receive antibiotics as well as those who did. Temperature data were Winsorized to correct for outliers. We also performed descriptive statistics for each group. Results There were 237,988 temperature readings for 1,568 ICU patients on CRRT. 1,153 patients received broad-spectrum antibiotics in ICU. In patients who received antibiotics in ICU and were presumed to have an infection, the mean temperature was 37.2°C prior to initiation of CRRT, 36.8°C while on CRRT, and 37.2°C following discontinuation of CRRT. In the 415 patients who did not receive IV antibiotics, the mean temperature was 36.9°C prior to initiation of CRRT, 36.6°C while on CRRT, and 37.0°C following discontinuation of CRRT. During each of the periods before, during, and after CRRT, patients who received antibiotics had significantly higher temperatures than those who did not (P < 0.001). Patients receiving antibiotics were generally younger (mean 60 years vs. 64 years, P < 0.001), had longer ICU stays (mean 29 days vs. 12 days, P < 0.001) and spent more time being ventilated (mean 23 days vs. 7 days, P < 0.001). The mean SOFA score on day one was similar (mean 11.1 in the antibiotic group and 10.5 in the other group). Conclusion This investigation suggests that patients have slightly lower temperatures while on CRRT, by on average less than half a degree. A similar effect is seen in both patients with infections as well as those without. Further work will be needed to determine what constitutes a true febrile response in this population. Disclosures All authors: No reported disclosures.

There are contrasting data on concomitant Impella device in cardiogenic shock patients treated with venoarterial extracorporeal membrane oxygenation (VA ECMO) (ECPELLA). This study sought to compare early mortality in patients with cardiogenic shock treated with ECPELLA in comparison to VA ECMO alone. We reviewed the published literature from 2000 to 2018 for randomized, cohort, case-control, and case series studies evaluating adult patients requiring VA ECMO for cardiogenic shock. Five retrospective observational studies, representing 425 patients, were included. Venoarterial extracorporeal membrane oxygenation with concomitant Impella strategy was used in 27% of the patients. Median age across studies varied between 51 and 63 years with 59–88% patients being male. Use of ECPELLA was associated with higher weaning from VA ECMO and bridging to permanent ventricular assist device or cardiac transplant in three and four studies, respectively. The studies showed moderate heterogeneity with possible publication bias. The two studies that accounted for differences in baseline characteristics between treatment groups reported lower 30 day mortality with ECPELLA versus VA ECMO. The remaining three studies did not adjust for potential confounding and were at high risk for selection bias. In conclusion, ECPELLA is being increasingly used as a strategy in patients with cardiogenic shock. Additional large, high-quality studies are needed to evaluate clinical outcomes with ECPELLA.

Patients with febrile neutropenia (FN) often are subject to antibiotic and diagnostic test overuse. We sought to improve appropriate use of antimicrobials and diagnostic tests for patients with FN. We used a blended quality approach with Lean Six Sigma tools and iterative improvement of a clinical decision aid to guide providers through empirical antimicrobial selection and diagnostic evaluation of patients with FN during a yearlong period. We evaluated the incidence of nonadherence to best practice before, during, and after rollout of a clinical decision aid in conjunction with an educational initiative. At baseline, 71% of patients with FN had at least one critical deviation from best practice. During the project, the percentage decreased to 27.3%; 4 months after the project was completed, the percentage was 33.3% ( P = .04). A clinical decision aid can improve adherence to best practices for the empirical management of FN.